Neural-Adaptive Stochastic Attitude Filter on SO(3)

Hashim, Hashim A.; Abouheaf, Mohammed I.; Vamvoudakis, Kyriakos G.

doi:10.1109/lcsys.2021.3123227

Cited by 5 publications

(6 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is worth noting that α b → 0 and ||α σ || → 0 as q → ∞. In view of the nonlinear dynamics in (27) and the expression in (28), one obtains…”

Section: B Error Dynamics and Nn Approximationmentioning

confidence: 98%

“…with W ∈ R q×m being a matrix of synaptic weights, ϕ(x) ∈ R q standing for an activation function, and α f ∈ R m describing an approximated error vector. The activation function can have high-order connection elements, for example, Gaussian functions, radial basis functions (RBFs) [38], and sigmoid functions [27,39]. Our strategy is to reach accurate attitude and gyro bias estimation as well as attenuate gyro noise stochasticity effect which, in turn, will result in accurate estimation of nonlinear attitude dynamics.…”

Section: B Error Dynamics and Nn Approximationmentioning

confidence: 99%

“…Over the past few years, adaptive NNs have been shown to be efficient for online estimation of unknown high-order nonlinear dynamics. Successful applications of NNs include but are not limited to nonlinear attitude filtering [27], adaptive consensus of networked systems [28], trajectory tracking of ground robots [29], stochastic nonlinear systems [27,30], and strict-feedback multi-input multi-output systems [31]. Accurate NN estimation of unknown high-order nonlinear dynamics results in a successful control process [5,28,29].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Adaptive Neural Network Stochastic-Filter-Based Controller for Attitude Tracking With Disturbance Rejection

Hashim

Vamvoudakis

2024

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

This paper proposes a real-time neural network (NN) stochastic filter-based controller on the Lie Group of the Special Orthogonal Group SO(3) as a novel approach to the attitude tracking problem. The introduced solution consists of two parts: a filter and a controller. Firstly, an adaptive NN-based stochastic filter is proposed that estimates attitude components and dynamics using measurements supplied by onboard sensors directly. The filter design accounts for measurement uncertainties inherent to the attitude dynamics, namely unknown bias and noise corrupting angular velocity measurements. The closed loop signals of the proposed NN-based stochastic filter have been shown to be semi-globally uniformly ultimately bounded (SGUUB). Secondly, a novel control law on SO(3) coupled with the proposed estimator is presented. The control law addresses unknown disturbances. In addition, the closed loop signals of the proposed filter-based controller have been shown to be SGUUB. The proposed approach offers robust tracking performance by supplying the required control signal given data extracted from low-cost inertial measurement units. While the filter-based controller is presented in continuous form, the discrete implementation is also presented. Additionally, the unit-quaternion form of the proposed approach is given. The effectiveness and robustness of the proposed filter-based controller is demonstrated using its discrete form and considering low sampling rate, high initialization error, high-level of measurement uncertainties, and unknown disturbances.

show abstract

“…It is worth noting that α b → 0 and ||α σ || → 0 as q → ∞. In view of the nonlinear dynamics in (27) and the expression in (28), one obtains…”

Section: B Error Dynamics and Nn Approximationmentioning

confidence: 98%

Section: B Error Dynamics and Nn Approximationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Neural Network Stochastic-Filter-Based Controller for Attitude Tracking With Disturbance Rejection

Hashim

Vamvoudakis

2024

IEEE Trans. Neural Netw. Learning Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…For the measurand x, there are two measurements, x A and x B , acquired through different approaches. In figure 1, the transfer functions for x A and x B are complementary to each other, and thus the CF in figure 1 is an all-pass filter (APF) for the measurand x, as described by equation (7) in the s-domain. In the case of MARG-based attitude estimation, a high-pass filter is applied to the gyroscope, in order to eliminate the accumulative error caused by the integration of angular velocity.…”

Section: Classic and Generalized Cfmentioning

confidence: 99%

“…The acronym MARG can also be intepreted as the observations of three different vectors, i.e. magnetic field, angular rate, and gravity [5][6][7].…”

Section: Introductionmentioning

confidence: 99%

Discrete-time complementary filter for attitude estimation based on MARG sensor

Li¹,

Xu²,

Tang³

et al. 2022

Meas. Sci. Technol.

View full text Add to dashboard Cite

MARG sensor, which stands for the combination of magnetometer, accelerometer, and gyroscope, is widely used for three-dimensional attitude measurement. Among the mainstream solutions for MARG-based attitude estimation, complementary filter (CF) is normally regarded as a simplified alternative to Kalman filter (KF), mainly because CF can save the amount of calculation. A dual-vector discrete-time complementary filter (DV-DTCF) and its tuning methods are introduced in this paper. Different from the quaternion-based attitude estimation algorithms, DV-DTCF has linear measurement model, since it utilizes the gravity and geomagnetic vectors as its state variables instead of quaternion. This feature of DV-DTCF can avoid linearization error or the use of nonlinear algorithms, and can also greatly reduce its computational complexity. More interestingly, it is analytically revealed, and is also experimentally proven, the proposed DV-DTCF is fully equivalent to a fixed-gain KF. This fascinating fact straightforwardly leads to the tuning methods of DV-DTCF via the corresponding fixed-gain KF and Riccati equation. Such tuning methods of DV-DTCF are based on the statistic characteristics of MARG sensor noise, and that makes them solid and feasible. According to experiment results, DV-DTCF can achieve the same accuracy as that of commonly used KF algorithms in MARG-based attitude estimation, but with much lower time consumption. Hence, the proposed DV-DTCF is especially suitable for the applications that have strict limitations on computational costs.

show abstract

An Ant Colony Optimization with Turn-Around-Time Reduction Mechanism for the Robust Aircraft Maintenance Routing Problem

Eltoukhy

Mostafa

2022

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Neural-Adaptive Stochastic Attitude Filter on SO(3)

Cited by 5 publications

References 22 publications

Adaptive Neural Network Stochastic-Filter-Based Controller for Attitude Tracking With Disturbance Rejection

Adaptive Neural Network Stochastic-Filter-Based Controller for Attitude Tracking With Disturbance Rejection

Discrete-time complementary filter for attitude estimation based on MARG sensor

An Ant Colony Optimization with Turn-Around-Time Reduction Mechanism for the Robust Aircraft Maintenance Routing Problem

Contact Info

Product

Resources

About